The Agent Skills Directory

[SAFE]: The skill provides instructional guidelines for agent autonomy in software development workflows. It defines a policy where agents proceed with reversible actions like code editing without immediate confirmation, while explicitly requiring human approval for irreversible actions such as data deletion or external messaging. No malicious patterns, obfuscation, or unauthorized data access were found.
[NO_CODE]: The skill consists only of instructional markdown and configuration files without any executable scripts or binaries, which significantly minimizes the security attack surface.

principle-never-block-on-the-human